Preparing data for the matrix update

Prior to the matrix value update, you need to perform the following steps for data preparation.

1.  Open a version file or a network with the supply valid at the time of acquiring the count data (Opening and saving files).

2.  Import the demand matrices that you would like to update into Visum (Importing matrix values into Visum matrices).

3.  Make sure that the input is provided as numerical (possibly user-defined) attributes of links, zones, screenlines, turns or main turns (if a main node has been defined), lines or stop areas (Managing attributes and Editing attribute values of network objects).

4.  For the variant TFlowFuzzy, allocate the empirically determined values s (from several counts of the same volume of links, zones, screenlines, turns or main turns, lines or stop areas or from several counts in the same zone) to a numerical (possibly user-defined) attribute of links, zones, screenlines, turns or main turns, lines or stop areas (Editing attribute values of network objects and Reading an attribute file).

Note: s indicates the difference of z at which the fuzzy membership function decreases to the value 0 (Fundamentals: Methodological basics of TFlowFuzzy).

5.  For the variant Least squares, allocate a numerical (possibly user-defined) attribute of links, zones, screenlines, turns or main turns, lines or stop areas with values for the weights (Editing attribute values of network objects and Reading an attribute file).

6.  Assign the out-dated demand matrices to the current supply (Setting and calculating the parameters for PrT assignment and Setting and calculating PuT assignment parameters).

The volumes of the links, screenlines, turns or main turns or lines and/or passengers boarding or alighting at stop areas as well as the origin and destination demand of zones are calculated and are the basis for the flow matrix calculated with TFlowFuzzy or Least squares (Updating demand matrix with TFlowFuzzy and Updating demand matrices using the least squares method).